Sector: law

How to Regulate Artificial Intelligence

Curated on Posted on by Stefaan Verhulst

Oren Etzioni in the New York Times: “…we should regulate the tangible impact of A.I. systems (for example, the safety of autonomous vehicles) rather than trying to define and rein in the amorphous and rapidly developing field of A.I.

I propose three rules for artificial intelligence systems that are inspired by, yet develop further, the “three laws of robotics” that the writer Isaac Asimov introduced in 1942: A robot may not injure a human being or, through inaction, allow a human being to come to harm; a robot must obey the orders given it by human beings, except when such orders would conflict with the previous law; and a robot must protect its own existence as long as such protection does not conflict with the previous two laws.

These three laws are elegant but ambiguous: What, exactly, constitutes harm when it comes to A.I.? I suggest a more concrete basis for avoiding A.I. harm, based on three rules of my own.

First, an A.I. system must be subject to the full gamut of laws that apply to its human operator. This rule would cover private, corporate and government systems. We don’t want A.I. to engage in cyberbullying, stock manipulation or terrorist threats; we don’t want the F.B.I. to release A.I. systems that entrap people into committing crimes. We don’t want autonomous vehicles that drive through red lights, or worse, A.I. weapons that violate international treaties.

Our common law should be amended so that we can’t claim that our A.I. system did something that we couldn’t understand or anticipate. Simply put, “My A.I. did it” should not excuse illegal behavior.

My second rule is that an A.I. system must clearly disclose that it is not human. As we have seen in the case of bots — computer programs that can engage in increasingly sophisticated dialogue with real people — society needs assurances that A.I. systems are clearly labeled as such. In 2016, a bot known as Jill Watson, which served as a teaching assistant for an online course at Georgia Tech, fooled students into thinking it was human. A more serious example is the widespread use of pro-Trump political bots on social media in the days leading up to the 2016 elections, according to researchers at Oxford….

My third rule is that an A.I. system cannot retain or disclose confidential information without explicit approval from the source of that information…(More)”

Algorithms in the Criminal Justice System: Assessing the Use of Risk Assessments in Sentencing

Curated on Posted on by Stefaan Verhulst

Priscilla Guo, Danielle Kehl, and Sam Kessler at Responsive Communities (Harvard): “In the summer of 2016, some unusual headlines began appearing in news outlets across the United States. “Secret Algorithms That Predict Future Criminals Get a Thumbs Up From the Wisconsin Supreme Court,” read one. Another declared: “There’s software used across the country to predict future criminals. And it’s biased against blacks.” These news stories (and others like them) drew attention to a previously obscure but fast-growing area in the field of criminal justice: the use of risk assessment software, powered by sophisticated and sometimes proprietary algorithms, to predict whether individual criminals are likely candidates for recidivism. In recent years, these programs have spread like wildfire throughout the American judicial system. They are now being used in a broad capacity, in areas ranging from pre-trial risk assessment to sentencing and probation hearings. This paper focuses on the latest—and perhaps most concerning—use of these risk assessment tools: their incorporation into the criminal sentencing process, a development which raises fundamental legal and ethical questions about fairness, accountability, and transparency. The goal is to provide an overview of these issues and offer a set of key considerations and questions for further research that can help local policymakers who are currently implementing or considering implementing similar systems. We start by putting this trend in context: the history of actuarial risk in the American legal system and the evolution of algorithmic risk assessments as the latest incarnation of a much broader trend. We go on to discuss how these tools are used in sentencing specifically and how that differs from other contexts like pre-trial risk assessment. We then delve into the legal and policy questions raised by the use of risk assessment software in sentencing decisions, including the potential for constitutional challenges under the Due Process and Equal Protection clauses of the Fourteenth Amendment. Finally, we summarize the challenges that these systems create for law and policymakers in the United States, and outline a series of possible best practices to ensure that these systems are deployed in a manner that promotes fairness, transparency, and accountability in the criminal justice system….(More)”.

Crowdsourcing the Charlottesville Investigation

Curated on Posted on by Stefaan Verhulst

Internet sleuths got to work, and by Monday morning they were naming names and calling for arrests.

The name of the helmeted man went viral after New York Daily News columnist Shaun King posted a series of photos on Twitter and Facebook that more clearly showed his face and connected him to photos from a Facebook account. “Neck moles gave it away,” King wrote in his posts, which were shared more than 77,000 times. But the name of the red-bearded assailant was less clear: some on Twitter claimed it was a Texas man who goes by a Nordic alias online. Others were sure it was a Michigan man who, according to Facebook, attended high school with other white nationalist demonstrators depicted in photos from Charlottesville.

After being contacted for comment by The Marshall Project, the Michigan man removed his Facebook page from public view.

Such speculation, especially when it is not conclusive, has created new challenges for law enforcement. There is the obvious risk of false identification. In 2013, internet users wrongly identified university student Sunil Tripathi as a suspect in the Boston marathon bombing, prompting the internet forum Reddit to issue an apology for fostering “online witch hunts.” Already, an Arkansas professor was misidentified as as a torch-bearing protester, though not a criminal suspect, at the Charlottesville rallies.

Beyond the cost to misidentified suspects, the crowdsourced identification of criminal suspects is both a benefit and burden to investigators.

“If someone says: ‘hey, I have a picture of someone assaulting another person, and committing a hate crime,’ that’s great,” said Sgt. Sean Whitcomb, the spokesman for the Seattle Police Department, which used social media to help identify the pilot of a drone that crashed into a 2015 Pride Parade. (The man was convicted in January.) “But saying, ‘I am pretty sure that this person is so and so’. Well, ‘pretty sure’ is not going to cut it.”

Still, credible information can help police establish probable cause, which means they can ask a judge to sign off on either a search warrant, an arrest warrant, or both….(More)“.

E-residency and blockchain

Curated on Posted on by Stefaan Verhulst

Clare Sullivan and Eric Burger in Computer Law & Security Review: “In December 2014, Estonia became the first nation to open its digital borders to enable anyone, anywhere in the world to apply to become an e-Resident. Estonian e-Residency is essentially a commercial initiative. The e-ID issued to Estonian e-Residents enables commercial activities with the public and private sectors. It does not provide citizenship in its traditional sense, and the e-ID provided to e-Residents is not a travel document. However, in many ways it is an international ‘passport’ to the virtual world. E-Residency is a profound change and the recent announcement that the Estonian government is now partnering with Bitnation to offer a public notary service to Estonian e-Residents based on blockchain technology is of significance. The application of blockchain to e-Residency has the potential to fundamentally change the way identity information is controlled and authenticated. This paper examines the legal, policy, and technical implications of this development….(More)”.

 

Democratic Resilience for a Populist Age

Curated on Posted on by Stefaan Verhulst

Helmut K. Anheier at Project Syndicate: “… many democracies are plagued by serious maladies – such as electoral gerrymandering, voter suppression, fraud and corruption, violations of the rule of law, and threats to judicial independence and press freedom – there is little agreement about which solutions should be pursued.

How to make our democracies more resilient, if not altogether immune, to anti-democratic threats is a central question of our time. …

Democratic resilience demands that citizens do more than bemoan deficiencies and passively await constitutional reform. It requires openness to change and innovation. Such changes may occur incrementally, but their aggregate effect can be immense…

Governments and citizens thus have a rich set of options – such as diversity quotas, automatic voter registration, and online referenda – for addressing democratic deficiencies. Moreover, there are measures that can also help citizens mount a defense of democracy against authoritarian assaults.

To that end, organizations can be created to channel protest and dissent into the democratic process, so that certain voices are not driven to the political fringe. And watchdog groups can oversee deliberative assemblies and co-governance efforts – such as participatory budgeting – to give citizens more direct access to decision-making. At the same time, core governance institutions, like central banks and electoral commissions, should be depoliticized, to prevent their capture by populist opportunists.

When properly applied, these measures can encourage consensus building and thwart special interests. Moreover, such policies can boost public trust and give citizens a greater sense of ownership vis-à-vis their government.

Of course, some political innovations that work in one context may cause real damage in another. Referenda, for example, are easily manipulated by demagogues. Assemblies can become gridlocked, and quotas can restrict voters’ choices. Fixing contemporary democracy will inevitably require experimentation and adaptation.

Still, recent research can help us along the way. The Governance Report 2017 has compiled a diverse list of democratic tools that can be applied in different contexts around the globe – by governments, policymakers, civil-society leaders, and citizens.

In his contribution to the report, German sociologist Claus Offe, Professor Emeritus of the Hertie School and Humboldt University identifies two fundamental priorities for all democracies. The first is to secure all citizens’ basic rights and ability to participate in civic life; the second is to provide a just and open society with opportunities for all citizens. As it happens, these two imperatives are linked: democratic government should be “of,” “by,” and for the people….(More)”.

Chicago police see less violent crime after using predictive code

Curated on Posted on by Stefaan Verhulst

Jon Fingas at Engadget: “Law enforcement has been trying predictive policing software for a while now, but how well does it work when it’s put to a tough test? Potentially very well, according to Chicago police. The city’s 7th District police reportthat their use of predictive algorithms helped reduce the number of shootings 39 percent year-over-year in the first 7 months of 2017, with murders dropping by 33 percent. Three other districts didn’t witness as dramatic a change, but they still saw 15 to 29 percent reductions in shootings and a corresponding 9 to 18 percent drop in murders.

It mainly comes down to knowing where and when to deploy officers. One of the tools used in the 7th District, HunchLab, blends crime statistics with socioeconomic data, weather info and business locations to determine where crimes are likely to happen. Other tools (such as the Strategic Subject’s List and ShotSpotter) look at gang affiliation, drug arrest history and gunfire detection sensors.

If the performance holds, It’ll suggest that predictive policing can save lives when crime rates are particularly high, as they have been on Chicago’s South Side. However, both the Chicago Police Department and academics are quick to stress that algorithms are just one part of a larger solution. Officers still have be present, and this doesn’t tackle the underlying issues that cause crime, such as limited access to education and a lack of economic opportunity. Still, any successful reduction in violence is bound to be appreciated….(More)”.

The Nudging Divide in the Digital Big Data Era

Curated on Posted on by Stefaan Verhulst

Julia M. Puaschunder in the International Robotics & Automation Journal: “Since the end of the 1970ies a wide range of psychological, economic and sociological laboratory and field experiments proved human beings deviating from rational choices and standard neo-classical profit maximization axioms to fail to explain how human actually behave. Behavioral economists proposed to nudge and wink citizens to make better choices for them with many different applications. While the motivation behind nudging appears as a noble endeavor to foster peoples’ lives around the world in very many different applications, the nudging approach raises questions of social hierarchy and class division. The motivating force of the nudgital society may open a gate of exploitation of the populace and – based on privacy infringements – stripping them involuntarily from their own decision power in the shadow of legally-permitted libertarian paternalism and under the cloak of the noble goal of welfare-improving global governance. Nudging enables nudgers to plunder the simple uneducated citizen, who is neither aware of the nudging strategies nor able to oversee the tactics used by the nudgers.

The nudgers are thereby legally protected by democratically assigned positions they hold or by outsourcing strategies used, in which social media plays a crucial rule. Social media forces are captured as unfolding a class dividing nudgital society, in which the provider of social communication tools can reap surplus value from the information shared of social media users. The social media provider thereby becomes a capitalist-industrialist, who benefits from the information shared by social media users, or so-called consumer-workers, who share private information in their wish to interact with friends and communicate to public. The social media capitalist-industrialist reaps surplus value from the social media consumer-workers’ information sharing, which stems from nudging social media users. For one, social media space can be sold to marketers who can constantly penetrate the consumer-worker in a subliminal way with advertisements. But also nudging occurs as the big data compiled about the social media consumer-worker can be resold to marketers and technocrats to draw inferences about consumer choices, contemporary market trends or individual personality cues used for governance control, such as, for instance, border protection and tax compliance purposes.

The law of motion of the nudging societies holds an unequal concentration of power of those who have access to compiled data and who abuse their position under the cloak of hidden persuasion and in the shadow of paternalism. In the nudgital society, information, education and differing social classes determine who the nudgers and who the nudged are. Humans end in different silos or bubbles that differ in who has power and control and who is deceived and being ruled. The owners of the means of governance are able to reap a surplus value in a hidden persuasion, protected by the legal vacuum to curb libertarian paternalism, in the moral shadow of the unnoticeable guidance and under the cloak of the presumption that some know what is more rational than others. All these features lead to an unprecedented contemporary class struggle between the nudgers (those who nudge) and the nudged (those who are nudged), who are divided by the implicit means of governance in the digital scenery. In this light, governing our common welfare through deceptive means and outsourced governance on social media appears critical. In combination with the underlying assumption of the nudgers knowing better what is right, just and fair within society, the digital age and social media tools hold potential unprecedented ethical challenges….(More)”

Algorithmic Transparency for the Smart City

Curated on Posted on by Stefaan Verhulst

Paper by Robert Brauneis and Ellen P. Goodman: “Emerging across many disciplines are questions about algorithmic ethics – about the values embedded in artificial intelligence and big data analytics that increasingly replace human decisionmaking. Many are concerned that an algorithmic society is too opaque to be accountable for its behavior. An individual can be denied parole or denied credit, fired or not hired for reasons she will never know and cannot be articulated. In the public sector, the opacity of algorithmic decisionmaking is particularly problematic both because governmental decisions may be especially weighty, and because democratically-elected governments bear special duties of accountability. Investigative journalists have recently exposed the dangerous impenetrability of algorithmic processes used in the criminal justice field – dangerous because the predictions they make can be both erroneous and unfair, with none the wiser.

We set out to test the limits of transparency around governmental deployment of big data analytics, focusing our investigation on local and state government use of predictive algorithms. It is here, in local government, that algorithmically-determined decisions can be most directly impactful. And it is here that stretched agencies are most likely to hand over the analytics to private vendors, which may make design and policy choices out of the sight of the client agencies, the public, or both. To see just how impenetrable the resulting “black box” algorithms are, we filed 42 open records requests in 23 states seeking essential information about six predictive algorithm programs. We selected the most widely-used and well-reviewed programs, including those developed by for-profit companies, nonprofits, and academic/private sector partnerships. The goal was to see if, using the open records process, we could discover what policy judgments these algorithms embody, and could evaluate their utility and fairness.

To do this work, we identified what meaningful “algorithmic transparency” entails. We found that in almost every case, it wasn’t provided. Over-broad assertions of trade secrecy were a problem. But contrary to conventional wisdom, they were not the biggest obstacle. It will not usually be necessary to release the code used to execute predictive models in order to dramatically increase transparency. We conclude that publicly-deployed algorithms will be sufficiently transparent only if (1) governments generate appropriate records about their objectives for algorithmic processes and subsequent implementation and validation; (2) government contractors reveal to the public agency sufficient information about how they developed the algorithm; and (3) public agencies and courts treat trade secrecy claims as the limited exception to public disclosure that the law requires. Although it would require a multi-stakeholder process to develop best practices for record generation and disclosure, we present what we believe are eight principal types of information that such records should ideally contain….(More)”.

Let the People Know the Facts: Can Government Information Removed from the Internet Be Reclaimed?

Curated on Posted on by Stefaan Verhulst

Paper by Susan Nevelow Mart: “…examines the legal bases of the public’s right to access government information, reviews the types of information that have recently been removed from the Internet, and analyzes the rationales given for the removals. She suggests that the concerted use of the Freedom of Information Act by public interest groups and their constituents is a possible method of returning the information to the Internet….(More)”.

Rage against the machines: is AI-powered government worth it?

Curated on Posted on by Stefaan Verhulst

Maëlle Gavet at the WEF: “…the Australian government’s new “data-driven profiling” trial for drug testing welfare recipients, to US law enforcement’s use of facial recognition technology and the deployment of proprietary software in sentencing in many US courts … almost by stealth and with remarkably little outcry, technology is transforming the way we are policed, categorized as citizens and, perhaps one day soon, governed. We are only in the earliest stages of so-called algorithmic regulation — intelligent machines deploying big data, machine learning and artificial intelligence (AI) to regulate human behaviour and enforce laws — but it already has profound implications for the relationship between private citizens and the state….

Some may herald this as democracy rebooted. In my view it represents nothing less than a threat to democracy itself — and deep scepticism should prevail. There are five major problems with bringing algorithms into the policy arena:

  1. Self-reinforcing bias…
  2. Vulnerability to attack…
  3. Who’s calling the shots?…
  4. Are governments up to it?…
  5. Algorithms don’t do nuance….

All the problems notwithstanding, there’s little doubt that AI-powered government of some kind will happen. So, how can we avoid it becoming the stuff of bad science fiction? To begin with, we should leverage AI to explore positive alternatives instead of just applying it to support traditional solutions to society’s perceived problems. Rather than simply finding and sending criminals to jail faster in order to protect the public, how about using AI to figure out the effectiveness of other potential solutions? Offering young adult literacy, numeracy and other skills might well represent a far superior and more cost-effective solution to crime than more aggressive law enforcement. Moreover, AI should always be used at a population level, rather than at the individual level, in order to avoid stigmatizing people on the basis of their history, their genes and where they live. The same goes for the more subtle, yet even more pervasive data-driven targeting by prospective employers, health insurers, credit card companies and mortgage providers. While the commercial imperative for AI-powered categorization is clear, when it targets individuals it amounts to profiling with the inevitable consequence that entire sections of society are locked out of opportunity….(More)”.

Proudly powered by WordPress